Journal of Vocational Behavior 86 (2015) 10–19

Contents lists available at ScienceDirect

Journal of Vocational Behavior

j ourna l homepage: www.e lsev ie r .com/ locate / jvb

The influence of social comparison on career decision-making:Vocational identity as a moderator and regret as a mediator☆

Xu Li 1, Zhi-Jin Hou⁎, Yin JiaSchool of Psychology, Beijing Normal University, No. 19 Xinjiekouwai St., Haidian Dist., Beijing 100875, China

a r t i c l e i n f o

☆ This study is funded by the National Key Technolog⁎ Corresponding author.

E-mail address: zhijinhou@163.com (Z.-J. Hou).1 Xu Li is nowa doctoral student in theDepartment of

MD20740, USA. The authorswould like to express theirland, College Park for their valuable comments for revis

http://dx.doi.org/10.1016/j.jvb.2014.10.0030001-8791/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

Article history:Received 18 September 2014

This study investigated the influence of social comparison on career choice certainty and its po-tential mechanisms: regret as a mediator and vocational identity as a moderator. Before the for-mal experiment, 30 pairs of vocational values representing typical conflicts in career decision-making for Chinese university students were obtained. The formal experiment adopted asingle-factor (social comparison VS no comparison) between-subject design with vocationalidentity as an independent covariate. Ninety-eight junior and senior undergraduate studentsand graduate students in a university in China were invited to participate in the computer-controlled experiment, which involved vocational identity assessment, social comparisonmanip-ulation, and analogue career-choice scenario tests. Path analysis showed that: (a) Social compar-ison significantly and negatively predicted career choice certainty; (b) Regret partially mediatedthe effect of social comparison on career choice certainty; and (c) Vocational identity did notmoderate the path between social comparison and regret, but significantly moderated the nega-tive effect of regret on career choice certainty. These results indicated that in the collectivistic Chi-nese culture, individuals' career development trajectoriesmay not be totally independent and aresubject to influences by other people's choices, while emotion of regret and vocational identitydevelopment all play significant roles in this intricate process.

© 2014 Elsevier Inc. All rights reserved.

Keywords:Social comparisonRegretVocational identityCareer choice certainty

1. Introduction

Career decision-making is an important branch of vocational psychology and has received much research attention, particularlywith respect to exploring factors that may influence how certain people are about their choices, i.e., their career choice certainty(Osipow, Carney,Winer, Yanico, & Koschier, 1976; Tracey & Darcy, 2002). Vocational development theories have identified a numberof variables thatmay influence an individual's career choice, including personal factors like one's interests and values (Holland, 1997),and contextual factors like the structure and process of original family (Whiston & Keller, 2004) and significant others (Gati, 2013).However, examining existing literature that explored the contextual factors, no study has investigated the impact of social comparisonwith peers on career decision-making and choice certainty, which seems to be a potentially significant contextual force in shapingindividual's career paths, especially in China which predominantly embraces a collectivistic culture (Chen, 2009). A specific illustra-tion follows.

ies R&D Program of China (2012BAI36B03).

Counseling, Higher Education and Special Education, College of Education, University ofMaryland, College Park,sincere thanks to Dr. DennisM. Kivlighan, Jr., Dr. Derek Iwamoto, and Dr. JonathanMohr in University ofMary-ing the manuscript.

11X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

1.1. Social comparison in collectivistic cultures

Social comparison was first proposed by Festinger (1954) as the tendency to evaluate one's opinions and abilities using othergroup members as references. Studies investigating the relationship between culture and individual's self-concept have suggesteda possible link between one's cultural background and inclination to engage in social comparison (Heine & Hamamura, 2007;White & Lehman, 2005). While individualistic culture entails a self-concept as independent, autonomous and self-contained, in col-lectivistic culture the prevalent view of self is interdependent, connected, and embedded in interpersonal relationships with others(Markus & Kitayama, 1991;White & Lehman, 2005). For individuals with such an interdependent self-concept, research has indicatedthat they may pay more attention to their social relationships and standing, thus are more sensitive to the social aspect of self underpublic scrutiny and evaluation (Heine & Hamamura, 2007). Therefore, it may be reasonable to infer that individuals in a collectivisticcultural context like China may be significantly influenced by social comparisons since their social self remains a salient part of theiroverall self-concept, which partly exists in relation with others (Chen, 2009).

In fact, some research has already demonstrated significant associations between social comparison seeking and one's cultural ori-entation (Chung &Mallery, 1999; Heine & Hamamura, 2007;White & Lehman, 2005). For example, Chung andMallery (1999) foundout that higher collectivism was significantly correlated with increased desire to make general and upward social comparisons (“up-ward” means comparing with others who are better off or superior). White and Lehman (2005) found that in comparison withEuropean Canadians (representing individualistic culture), Asian Canadians (representing collectivistic culture) demonstrated signif-icantly higher tendency to seek social comparison, particularly upward comparisons that allow for self-improvement. Because theChinese culture is primarily characterized by collectivism where people adopt a self-in-relation perspective (Chen, 2009), it mightbe anticipated that social comparison would be prevalent in and have influence on Chinese people's daily lives. More specifically incareer decision-making, it may be a contextual factor that has potential salient roles but has not been addressed theoretically or ex-amined empirically in the Chinese culture.

1.2. Career choice certainty, decision-making, and social comparison

Although no previous research has directly examined the relationship between social comparison and career decision-making, anumber of studies have investigated the effects of social comparison on general decision making processes and outcomes (He & Bai,1997; Hoelzl & Loewenstein, 2005; Kumar, 2004; Linde & Sonnemans, 2012). For instance, in Kumar (2004), participants who weretold their friends had chosen their forgone alternative and received a greater discount tended to report less intention to stick tothis purchase choice, which indicated an inaction inertia resulting from social comparison and counterfactual regret. Hoelzl andLoewenstein (2005) pointed out that social counterfactuals increased individuals' tendency to stick with an investment. Within theChinese context, though limited in number and scope, research results still indicated that social comparison influenced the risk aver-sion of people who had experienced gain in a prior choice (He & Bai, 1997). Above evidence implied that social comparison affectedpeople's choice behaviors by either increasing or decreasing theirwillingness/intention to continue their former choice. Therefore, weinfer that comparing with others may potentially influence individuals' choice certainty: how certain they feel about and howmuchthey hold on to their previous choices are significantly shaped by the messages obtained from comparing with others.

Career choice is a very important type of choice people make in their lives, and the aforementioned findings have potential impli-cations for career decision research as well. These findings evidence that social comparison influences choice certainty in generaldecision-making, thus suggest that such effects may also apply to people's choice certainty when making a career decision. In thefield of vocational psychology, career choice certainty is an important variable that conceptualizes an individual's certainty and com-mitment to a career choice (Tracey & Darcy, 2002). Moreover, it has direct effects on career indecision (Osipow et al., 1976), careerchoice satisfaction (Robinson & Cooper, 1998), and turnover behavior (McLean, Bryan, Tanner, & Smits, 1993). However, in existingvocational psychology studies that examine what variables predict career choice certainty as an important career outcome, the con-textual variable of social comparisonhas not been addressed, even though as aforementioned, research in other disciplines of psychol-ogy has already suggested such a link.

Several researchfindings about the characteristics of Chinese social context and Chinese students' career choice have also indirect-ly implied this prediction relationship specifically in the Chinese population. In the high-context Chinese culture (Kim, Pan, & Park,1998), career decisions are often made within a larger context where family, peers, or social factors may play salient roles aboveand beyond one's individual circumstances (Li, Hou, & Feng, 2013). This suggests that contextual factors like family's career expecta-tion or other peers' career choice may have potential significant effects on Chinese students' career decision-making. WhileGottfredson (1996) proposed three aspects of career choice (i.e. sex type, prestige, and interest) in her circumscription and compro-misemodel, for Chinese university students social prestigewas always considered as themost important factor in situations calling forcompromise, and Chinese people tend to associate one's occupationwith his or her social status (Zhou&Ma, 2007).Moreover, studiesinvestigating Chinese parents' career expectation on their children have found achievements and social status related expectations tobe its most salient facet (Hou, Chen, Zhou, & Li, 2012). Given that people in collectivistic cultures have a greater tendency tomake up-ward comparisons in terms of ability or social standing (White & Lehman, 2005), and given the prevalent emphasis on social statusand prestige in career-related aspects in the Chinese society, it might be expected that Chinese students would tend to comparewith others in terms of career choice which is widely perceived to reflect their social status, and this act of social comparison mayhave a potential impact on how certain people feel about their earlier career choices.

Taken together, the first objective of this study is to investigatewhether the predicting effect of social comparison on career choicecertainty could be found in Chinese university students. This research question is expected to expand current literature by examining

12 X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

the culturally relevant variable of social comparison as a specific contextual factor in influencing career decision-making, which hasnot been explicitly addressed in vocational theories or empirical studies, especially with regard to the Chinese population.

1.3. Regret as a potential mediator

Reflecting on earlymodels of career choice, scholars have noted that theportrayal of career development and choice as a complete-ly linear and rational process has left out the significant roles of intuition and emotion in shaping individual's career choice (Kidd,1998; Krieshok, Black, & McKay, 2009). Though some recent models (e.g., Saka & Gati, 2007) started to incorporate emotions intothe understanding of career decision-making, no study has examined (1) do emotions have significant effects and (2) what specificroles/effects emotions have in individual's career choice process, especially with an experimental design.

From a broader perspective, however, scholars investigating general decision making have found support for the significant rolesof emotion in people's decision-making process, especially regret, which is a negative emotion generated by comparing between ac-tual and expected (better) outcomes (Zeelenberg, 1999). On one hand, research has supported the association between regret andsocial comparison: people scored high on social comparison orientation reported more regret than participants with a low need tocompare (White, Langer, Yariv, & Welch, 2006). On the other hand, regret is shown to shape people's decision-making behaviors.For example, Bailey and Kinerson (2005) reported that the experience of regret with a particular type of investment did reduceone's tendency tomake a similar investment. Schwartz et al. (2002) found that maximizers (who examine every possible alternativesand search for the best) reported more engagement in social comparison and more regret than satisficers (who choose one exceedsthe acceptability threshold) inmaking purchase decisions, and regret played a partialmediational role betweenmaximization and af-fect (depression and happiness).

In understanding these effects, two important theoretical concepts have been proposed: counterfactual comparison as a necessaryantecedent of regret (Zeelenberg, 1999), and Affect-As-Information as the mechanism through which regret influences choice(Schwarz & Clore, 1988).When an individual engages in social comparisonwith another person who is in the same situation but bet-ter off, the “what if I had chosen that” or “what if I were him in the situation” thoughts actually provide the counterfactual comparisonswhichmight trigger the emotion of regret. This feeling then will be a piece of necessary information which he or she uses when eval-uating and deciding on the choices, as theorized in the Affect-As-Information theory (Schwarz & Clore, 1988).

Therefore, it might be speculated thatmaking social comparisonwith someonewho chose a forgone attractive career choicemightserve as the counterfactual information that triggers the emotion of regret, which in turn influences the certainty one feels about theiroriginal career choice. The second objective of this research, then, is to explore the potential mediating effect of regret between socialcomparison and career choice certainty.

1.4. Vocational identity as a potential moderator

The third objective of this study is to investigate the potential moderators that could possibly buffer the effects as hypothesizedabove.When studying general self-identity, Marcia (1966) concluded that people with higher self-identity tended to havemore crys-tallized self-concept, more stable self-evaluation and greater decidedness. Holland, Gottfredson, and Power (1980) operationalizedvocational identity as a specific component of one's general self-identity, which indicates “the possession of a clear and stable pictureof one's goals, interests and talents (p. 1191)”. Porfeli, Lee, Vondracek, and Weigold (2011) pointed out that vocational identityreflected the status and level of an individual's understanding of him- or herself in terms of career development based on cumulativeexploratory experiences. Thereby, it could be speculated that people with a higher level of vocational identitywould bemore likely tohave a clearer understanding of their own career values and goals. While social comparison may carry external information that po-tentially “disturbs” one's original choice, people with higher identity are thus hypothesized to be less susceptible to such impacts andhave greater certainty and commitment to their choices.

(a) (b)

Fig. 1. Relationships between social comparison, regret, vocational identity, and career choice certainty. (a) Hypothetical Model of All Investigated Variables;(b) Estimated Overall Model. To generate product terms, vocational identity and regret scores are grand centered. *p b .05. **p b .01. ***p b .001.

13X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

Although few empirical studies have directly tested themoderation of vocational identity in career choice certainty, some researchlends support to its relevance to career decision-making in general. For example, vocational identity has been consistently found tonegatively correlate with career indecision or indecisiveness (Fretz & Leong, 1982) and positively correlate with career decidednessand choice certainty (Blustein, Devenis, & Kidney, 1989). Given the above theoretical rationale and the empirical evidence of its pro-tective role, the third objective of this study is to investigate the moderation effect of vocational identity in buffering the impact ofsocial comparison on career choice certainty.

1.5. Summary of research questions and hypotheses

In sum, this study tests three hypotheses. Hypothesis 1 (H1) posits the existence of a significant negative relationship between so-cial comparison and Chinese university students' career choice certainty. Hypothesis 2 (H2) predicts that emotion of regret will serveas a mediator between social comparison and career choice certainty. Specifically, engaging in threatening social comparison wouldlead to greater regret over a prior career choice thus decreasing his/her choice certainty. Hypothesis 3 (H3) postulates that vocationalidentity will buffer (moderate) the prediction of social comparison on regret and career choice certainty, and regret on career choicecertainty. Specifically, higher vocational identity will attenuate all the above effects: when individuals with higher vocational identityengage in social comparison, in comparison to those with lower vocational identity, they will report less regret, and greater careerchoice certainty.

These three hypotheses are depicted in Fig. 1(a). Themain effect of vocational identity on career choice is not a focus of the currentstudy but this relationship is suggested by previous research, thus is retained in themodel. It's expected that this studywill add to ourexisting knowledge of career decision-making primarily by (1) incorporating an important yet unexplored contextual factor of socialcomparisonwhichmay bear particular significance in the Chinese cultural context, and (2) empirically examine the role of emotion incareer choice as a mediation mechanism of the above prediction effect in an experimental design.

2. Methods

2.1. Participants

Advertisementswere posted on the online forumof a university in Beijing and 98 undergraduate and graduate students (38malesand 60 females) were recruited to participate in this study. Their age ranged from 19 to 28 (M= 23.43, SD= 1.94), and their gradedistributionwas: 20 junior undergraduates, 17 senior undergraduates, and 23 first year, 16 second year and 22 third year master stu-dents. We decided to recruit upper-class undergraduate students (juniors and seniors) or graduate students to participate in thisstudy, because making a career decision is one of the primary career development tasks for this age group (Super, 1980), and it'smore immediate for them to complete this task in comparison to lower-class undergraduates (freshmen and sophomores) whohave just entered university.

2.2. Measures

2.2.1. Vocational identityThe vocational identity subscale in the My Vocational Status (MVS, Holland et al., 1980) inventory was used to measure the voca-

tional identity in this study. This subscale is unidimensional and contains 18 items, to which participants are required to respond on aYes–No forced choice format. Then the count of one's “No” response is regarded as the final score of vocational identity, with higherscores indicating higher level of identity and vice versa. Holland et al. (1980) reported good internal consistency of this overall scale(from .86 to .89), and in this study, the consistency coefficient was .82, which revealed its satisfactory reliability. Evidence supportingthe validity of this measure was reviewed and summarized in Holland, Johnston, and Asama (1993). These authors reported that vo-cational identity was related to pertinent constructs like ego identity, career indecision and indecisiveness, career beliefs, and occupa-tional values. Hou (2002) used MVS with a sample of Chinese students and revealed its reliability coefficient of .79. As expected, theMVS score negatively predicted career decision-making difficulty, which lent support to the validity of this scale for the Chinesesample.

2.3. Design of the formal experiment

The formal experiment of this study adopted a single-factor (social comparison) between-subject design. By experimental manip-ulation, participants were randomly assigned to two groups with equal probabilities of 50%. Those in the experimental group wereprimed to compare with a friendwhomade a different choice in similar scenario and obtained the corresponding desirable outcome,while those in the control group engaged in no social comparison. The other independent variable of vocational identity was assessedusing MVS as a continuous covariate. The whole process of formal experiment was controlled by a computer program coded usingVisual Basic language.

14 X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

2.4. Procedure

2.4.1. Part one: establishing a pool of work value pairsThe purpose of this part was to establish a pool of work value pairs that might capture the basic work value conflicts for Chinese

university students for use in constructing a conflicting job decision scenario for the formal experiment (Part Two). Sixteen corework values (see Appendix A) which Chinese university students consider the most when making career decisions were adoptedfrom Jin and Li (2005) and paired up to form 120 value dyads. Jin and Li (2005) used open-ended questionnaire and in-depth inter-view methods to obtain Chinese university students' narratives about their work values, and then used qualitative coding to extractspecific items to describe these work values. Twenty-four graduate students whomajored in career development and are also careercounselors in this university were invited to rate these 120 value pairs. They were asked to assess on an eight-point Likert scale howtypical each value pair represents a conflictual scenario (when a person has to compromise one for satisfying the other) for Chineseuniversity students. The 30 value pairs receiving the highest average scores (see Appendix B) were retained for use in the formalexperiment.

2.4.2. Part two: formal experimentThe design of the formal experiment was similar to Kumar (2004), which could be divided into three consecutive sections.In the first section, participants were asked to input their basic demographic information, including their gender, age, major and

grade. Afterwards, they were administered the MVS.In the second section, participants were presented with the 30 work value pairs in random order and they were asked to choose

one of the twowork values thatwas “comparativelymore important” to themand indicate the difficulty level inmaking this choice ona Likert scale from 1 (not difficult at all) to 8 (extremely difficult). Their reaction timewas also recorded for each pair. Then the com-puter program selected the pair that received the highest difficulty rating for use in the third section. If two ormore pairs had the samehighest rating, the onewith the longest reaction timewould be selected (Drugowitsch, Moreno-Bote, Churchland, Shadlen, & Pouget,2012). This procedure aimed at identifying the most “conflict-arousing” value pair for each individual, thus enhancing their identifi-cation and commitment to the following career choice scenario.

The third section consisted of the core part of the formal experiment: a career choice test which is an analogue of a real-life situ-ation that Chinese students often encounter. In the first step the computer program took the most “conflict-arousing” value pair (forexample, “high salary” VS “job matches interests and personality”) obtained in the second section to create a conflictual job selectionscenario. In this scenario, the participant had to choose between two job offers: Job A scored 7 on an 8-point scale in satisfying onevalue and 2 in satisfying its competing value, vice versa for Job B. Under this job description, an unscaled line segment was presentedto the participant with the left end representing Job A (coordinate set to be 0) and the right end representing Job B (coordinate set tobe 1). The participantwas instructed to click on any position (coordinate recorded as x1) along this line segment to indicate his or herdegree of preference between these two jobs (e.g., the closer one clicks towards Job A, the more one prefers Job A). Participants re-ceived computer feedback about which job they preferred after their response.

After this step, participants were randomly assigned to either the experiment or the control group. Those in the experiment groupwould be presented with materials aiming at priming their social comparison process, which asked the participant to imagine a peerfriend who has faced the same job selection scenario but made a different choice. Then the participant was instructed to write downthree advantages or good things that his/her peer friend could be able to enjoy due to her decision in a fixed format of “Because of his/her choice, he/she is better able to (Type in answers here)”. For the control group, some readingmaterial of the same length unrelatedto the research topic was displayed and they were asked to answer three questions after reading.

In the third step, the regret of participants in both groups was assessed similarly as in Kumar (2004). They were asked to indicatetheir level of regret when reflecting on the choice he or she just made on an eight-point Likert scale with 1 = not regretful at all and8 = extremely regretful.

Lastly, participantswere instructed to indicate their preference in the previous job selection scenario a second timewithout know-ing the exact position they had clicked on at the first time. The coordinate of the second choice x2 was recorded as well. The rationalefor having participants rate their preferences a second time without knowing their prior ratings is that it attends more to theirperceptual level, and minimizes the impact of a variety of possible cognitive processes which may also affect how they reporttheir job preference. A more specific discussion of this operationalization and its implication will be presented in the Discussionsection.

After completing all these procedures, participants were given small souvenirs for their participation, thanked and dismissed.

2.5. Operationalization of investigated variables

The four investigated variables in this studywere operationalized as below. Social comparisonwas represented by the experimentmanipulation. Vocational identity was operationalized as participant's total score on theMVS scale. Regret was operationally definedas the regret score the participant reported during the experiment. And career choice certainty (Cert) was operationalized as the dif-ference between the two preference coordinates, specifically, if x1≤ .5, then Cert= x1− x2; if x1 N .5, then Cert= x2− x1. According tothis operational definition, Cert ranged from−1 to 1, with a smaller value representing a greater shift away from one's prior choice,thus a lower career choice certainty.

15X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

3. Results

3.1. Preliminary analysis

Before examining the hypothesized model, some preliminary tests were conducted to ensure the validity of subsequent analyses.Firstly, the homogeneity of experiment group and control groupwas examined. This was done by testingwhether the differences be-tween these two groupswere significant on all variables tested before the experimentmanipulation, i.e., gender, age, grade, vocationalidentity score, the first preference coordinate x1. Results indicated that none of these differences were significant (Table 1), thus lend-ing empirical support to the homogeneity of these two randomly assigned groups.

The second set of tests was to examinewhether demographic variables might have significant effects on the four investigated var-iables. Specifically, we inspected whether the vocational identity score, the first and second preference coordinates, the emotion ofregret, and career choice certainty differed across gender and grade. No significant ANOVA results were obtained, showing that thedemographic variables included in this study had no significant effects on the four research variables. They were thus not includedin subsequent analyses.

3.2. Formal analysis: testing the hypothesized model

The descriptive statistics are displayed in Table 1. Path analysis was adopted to collectively test the whole model. The variables ofsocial comparison (categorical variable representing experimental manipulation), vocational identity (grand centered), regret (grandcentered), career choice certainty, the product terms of social comparison × vocational identity and regret × vocational identity weresimultaneously entered into the Mplus 6.0 software. The obtained final model was depicted in Fig. 1(b).

The overall model fit was excellent: χ2 = .080, p= .96 N .05. RMSEA= .000, CFI = 1.000, and SRMR= .006. Inspecting the pathcoefficients, it could be found that social comparison had significant and negative relationship to career choice certainty (β=− .160,p b .05), and significant and positive relationship to regret (β= .271, p b .01), while regret significantly and negatively related to ca-reer choice certainty (β = − .523, p b .001). These three paths supported the Hypothesis 1 proposed in this study. To evaluate themediation effect of regret between social comparison and career choice certainty, bootstrap technique was adopted due to its highestpower and best type I error control (Hayes, 2009). Results indicated a significant partialmediation effect of regret between social com-parison and career choice certainty: standardized total indirect effect=− .142, SE= .056, p b .05. The proportion of mediation effectin the overall effect was− .142/− .302 = 47.0%. These results lent support to Hypothesis 2.

Examining the variable of vocational identity, its main effect on career choice certainty was not significant (β= .107, p N .10). Asfor the product terms, the interaction of vocational identity and social comparison failed to predict regret (β=− .169, p= .073 N .05)or career choice certainty (β=− .115, p N .10). The interaction between vocational identity and regret significantly predicted careerchoice certainty (β = .215, p b .01). These results indicated that, although vocational identity did not moderate the direct effect ofsocial comparison on career choice certainty, it did manifest significant buffering effects on the indirect paths: the higher vocationalidentity, the lower impact of regret on career choice certainty. Therefore, Hypothesis 3 was partly supported.

4. Discussion

4.1. Social comparison and career decision-making

The core question of this study was to investigate whether social comparison would influence Chinese university students' careerchoice certainty.With a between-subject experimental design, results supported such a causal relationship as proposed in Hypothesis1.When a student received information involving a comparison with a peer who hadmade a different choice highly valued by them-selves as well, and was primed to think about the advantages the peer friend enjoyed, the student's certainty about their original ca-reer choice was significantly decreased.

Table 1Descriptive statistics and difference tests between experimental and control group.

All (N = 98)M (SD)

Experimental (n = 52)M (SD)

Control (n = 46)M (SD)

Test statistic

MVS total score 9.959 (4.225) 9.690 (3.883) 10.260 (4.606) Z = − .528x1 .440 (.239) .447 (.220) .432 (.262) F = .089Regret 2.898 (1.640) 3.330 (1.665) 2.410 (1.484) F = 8.134**x2 .467 (.245) .507 (.207) .422 (.278) Z = −1.630Choice certainty − .039 (.143) − .084 (.174) .012 (.072) Z = −2.606**

Note. Test statistics examined whether significant differences existed between the experimental and control groups. x1 = coordinate of preference before manipula-tion; x2 = coordinate of preference after manipulation. On MVS total score, x2, and choice certainty, the two groups did not have homogeneous variance, so thenon-parametric Mann–Whitney U Test was conducted. *p b .05. **p b .01. ***p b .001.

16 X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

This result seemed to be the first direct empirical evidence supporting a causal effect of social comparison on careerchoice, and suggested that individuals' career choices might not be mutually independent. How one chose his or her careerpath was not only based on personal preferences and abilities but might be also influenced by the choices of people aroundhim/her like peers as found in this study. Therefore, it would seem to be of additional value to incorporate social comparisoninto current career decision-making theories or models, and view one's career choice behaviors as both individually and so-cially based.

A further look at the data raised another interesting point. If we took the middle point of the line segment in the job choice task(coordinate = .5) as a threshold differentiating whether one had chosen Job A or B, it could be discovered that 85 out of the 98 par-ticipants (86.7%) did notmanifestly change their choice after experimental manipulation (i.e., from Job A to Job B or vice versa).Whilepracticing career service to Chinese university students, the authors have had some observations about students' cognitive processesthat underlie similar comparison-related career choice difficulties. For example, some students thought that it was not good to com-pare and vacillate whenmaking decisions and to not commit to the choice one had justmade; others indicated even if another choiceweremade, theymight also have other regrets aswell so there was no point in changing the decision; yet others felt that the decisionprocess was so difficult and emotionally taxing thus they would rather choose to settle for a choice. From these observations, it couldbe seen that factors like social desirability (the first response), cognitive dissonance (the second response) and avoidance of negativeemotions (the third response) and so on might all be at play to mitigate the impact of external disturbances brought about by socialcomparison and due to these reasons, receiving one piece of such information would usually not lead to completely changing thechoice on the cognitive level.

Interestingly however, results of this study showed that people primedwith social comparisonwere significantly “disturbed” fromtheir original preference among these two job options. By asking participants to indicate their job preference along an unscaled linesegment without seeing their prior rating, this experimental design tapped more into the perceptual rather than cognitive level,where thedisturbing effect of social comparisonwas revealed to be salient. Gati (2013) highlighted the trend of studying career choicefrom focusing only on conscious cognitive processes to also including unconscious intuitive processes, arguing that the role of the lat-ter is important and complementary to the conscious cognitive decision-making. The finding in this study suggested that contextualfactors (social comparison) did affect individual's career choice certainty on the unconscious perceptual level. Since people's careerchoices are usually not made instantaneously, but may go through a series of stages like intention generation, maintenance andthen the final execution (Kuhl, 1992), the results in this study highlighted the possibility that even if engaging in one social compar-ison scenario did not directly change one's decision on a conscious cognitive level, cumulative exposure to such comparison scenariosin life might create cumulative “unconscious disturbances” great enough to alter one's career choice.

4.2. The mediation role of regret

In spite of the literature that has extensively discussed the role of emotion in general decision making (Peters, Vastfjall, Garling, &Slovic, 2006), the specific effects of emotion in the realm of vocational psychology did not seem to have received sufficient researchattention (Kidd, 1998). By incorporating regret inHypothesis 2, this study provided preliminary empirical evidence onhowemotion isinterweaved into individual's career decision-making processes.

Results supported Hypothesis 2. Regret exhibited significant main effect in predicting career choice certainty. This seemed to besuggesting that, as a specific part of human's general decision-making, career decision was indeed an emotion-laden process (Kidd,1998). Early career theorists tended to conceptualize career development and decision-making to be a completely rational process(e.g., Parsons, 1909) based on unbiased hypotheses testing (Osipow, 1983), yet recent theorists argued that our rationality wasbounded and the career decisions might also incorporate abundant of emotional, intuitional and unconscious elements(Kahneman, 2003; Krieshok et al., 2009; Saka & Gati, 2007). The results in this study seemed to lend direct empirical support to theperspectives of recent theorists: career decision-making might be better construed as an interactive process including both cognitiveand emotional components, instead of being done in a completely rational and mechanical manner.

Regret also partiallymediated the negative effect of social comparison on career choice certainty, whichwas consistentwith otherstudies investigating the specific mechanisms of emotion in more general decision-making processes (Hoelzl & Loewenstein, 2005;Kumar, 2004). The mediation model could be understood within the general framework of Affect-as-Information theory (Schwarz& Clore, 1988), which asserted that the anticipated or previously experienced emotions towards an option could be important infor-mation that individuals refer towhenmaking a decision. In the experiment, participants were primed to imagine their peers who hadmade different choices in the similar situation andwere better off. This social comparisonwas also the counterfactual comparison thatwas found to be the prerequisite for the generation of regret (Zeelenberg, 1999). Therefore, the emergence of the mediation modelsuggested that social comparison as a form of counterfactual comparison elicited a feeling of regret in individuals about their earlierdecision, while the emotion of regret, as important reference information, further led to a significant decline in their choice certainty.Mediation effect analysis showed that the mediation effect accounted for 47.0% of the overall effect, which suggested that nearly halfof the influence of social comparison on career decision-making was through the path of regret, lending support to the salient role ofemotion in the career realm.

4.3. The moderation role of vocational identity

The third research objective aimed at investigating themoderationmechanism in the overall model, and results, in partial supportof the corresponding Hypothesis 3, only revealed significantmoderation effect of vocational identity on the path from regret to career

17X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

choice certainty. Previous research found that people with higher general self-identity tended to have more crystallized self-concept,more stable self-evaluation and greater decidedness (Marcia, 1966), and the significantmoderation effect of vocational identity in thisstudy seemed to have expanded this general finding into the specific field of career development. Specifically, higher vocational iden-tity, i.e., a higher level of clarity and certainty about one's career interests, values, and aspirations, would be a protective factor whenone's career choicewas disturbed byother people's career paths and outcomes via the process of social comparison. Its non-significantmoderation between social comparison and regret but significantmoderation between regret and career choice certainty further sug-gested that, in face of threatening social comparison information, although the possession of high vocational identity might not be ef-fective in preventing one from feeling the initial emotional reaction of regret, it nevertheless significantly helped to regulate the regretand to mitigate its negative impact on subsequent career choice actions. Taken together, these results indicated that an individual'svocational identity level helped tomaintain his/her decidedness and commitment to an earlier choice under the possible disturbancescoming from a broader social context, and it was the interaction of the vocational identity as an individual factor and the social com-parison as a contextual factor that conjointly affected the career decision-making outcomes.

In spite of the significant moderation effect, the main effect of vocational identity on career decision-making was not supported.This appeared to be contradictory to previous studies (e.g., Blustein et al., 1989) where vocational identity was correlated with out-come variables like career commitment. However, a further inspection indicated that previous studies were all based on self-reportscales, whichmeasured the status of vocational identity and career decision across a certain length of time that tended to bemore sta-ble; while in this experimental research, the career choice certainty actually reflected how certain and committed participants wereafter the disturbance of social comparison within a short period of time. Combining these observations, it could be tentatively spec-ulated that vocational identity could directly predict individual's career decision-making status across a comparatively longer periodof time, while during the short disturbance imposed on individual by his/her context, vocational identity would primarily display abuffering effect. This hypothetical speculation might be tested in future research.

5. Implication for career guidance and counseling

Firstly, the significant direct and indirect effects of social comparison on career choice certainty found on a sample of Chinese stu-dents highlighted the need to contextualize them in their social environmentwhen providing career service. Career counselorsmightfind it helpful to explore the possible contextual influences like other peers' career decisions and paths, and their potential impact onthe client. This study also provided support to the claim that career decision-makingmight be an emotion-laden process (Kidd, 1998).Therefore, emotion work seemed also a salient part of career guidance or counseling. The significant mediation effect implied that, ifcareer counselor could effectively work on emotions associatedwith client's career choice (e.g. helping the client express and explorerelevant emotions), the disturbance of contextual influence on the client's career choice could be alleviated. Lastly, the salient buffer-ing effects of vocational identity might help shed light on the importance of helping clients' identity development by encouragingthem to engage in career exploration and reflection (Porfeli et al., 2011). With greater awareness of one's own interests and aspira-tions, an individual would potentially be less susceptible to external disturbances like social comparison, and when the emotion ofregret is inevitably aroused, onewould be better able to regulate this emotion so that it has less negative impact on subsequent careerchoices.

6. Limitation and future research directions

Firstly, social comparison is a variable that has been demonstrated to correlate with specific cultural contexts (Chung & Mallery,1999). However, by design this study only investigated how social comparison influenced career decision-making of universitystudents in the collectivistic Chinese culture.Without specifically including the cultural variable andmaking a cross-cultural compar-ison, it would be difficult to determine whether social comparison will similarly influence career choice in other cultures(e.g., individualistic cultures) and whether cultural contexts can explain possible differences. Therefore, future studies may be wellwarranted to conduct a multicultural study by adopting similar paradigms on samples from different cultures and examine the pos-sible cultural differences.

In operationalizing social comparison, this study only manipulated participants to engage in threatening peer comparisonscenarios. However, social comparison can be both upward and downward and with many other targets (Festinger, 1954),and these more nuanced investigations were not included in this study. Future research may complement the current studyby investigating how different types of social comparison with different targets may influence individuals' career choice morespecifically.

Thirdly, theMVSused in this studywasbased onHolland (1980)'s unidimensional conceptualization and only captured the aware-ness and clarity components of vocational identity. This measure may not comprehensively and dynamically reflect vocational iden-tity as other measures do (e.g., Porfeli et al., 2011). There was also conceptual overlap between MVS and career decision certainty,which may be confounded with the dependent variable. Future research may use other measures of vocational identity to test ifthe moderation effect can be replicated.

Lastly, the participants in this study were recruited from one university in Beijing. Although homogeneous sample could bettercontrol irrelevant factors in the experiment, it compromises the generalizability of research findings. Future research may expandthe sample to different populations so that the current model could be tested or revised.

18 X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

Appendix A. Sixteen core career values of Chinese university students (adopted and revised from Jin & Li, 2005)

Value # Career value

1 High salary2 High social status3 Good prospect of promotion4 Stable and secure job5 Job-major match6 Working in big companies7 Good working environment8 Job matches interests and personality9 Taking care of parents10 Good for marriage and my family11 Creative work12 Good interpersonal atmosphere13 Job with high social approval14 Work independently and autonomously15 Job manifests my values16 Good employee benefits

Appendix B. Thirty value pairs with highest average conflict ratings used in this study

Pair # Value 1 Value 2 Mean Conflict Rating

1 High salary Taking care of parents 6.3752 High salary Job matches interests and personality 6.1903 Job matches interests and personality Taking care of parents 5.5334 Good prospect of promotion Job matches interests and personality 5.4675 High salary Stable and secure job 5.3816 High salary Job manifests my values 5.2507 Stable and secure job Job manifests my values 5.2228 High social status Work independently and autonomously 5.2119 High salary Job-major match 5.16710 High salary Good prospect of promotion 5.16711 High salary Good interpersonal atmosphere 5.07112 High salary Work independently and autonomously 5.05313 High social status Job matches interests and personality 5.00014 Stable and secure job Work independently and autonomously 4.94115 High salary Creative work 4.92316 Good prospect of promotion Taking care of parents 4.91717 High social status Good interpersonal atmosphere 4.87518 Stable and secure job Creative work 4.85019 Job matches interests and personality Good for marriage and my family 4.83320 Taking care of parents Creative work 4.83321 Good for marriage and my family Job manifests my values 4.83322 High salary Good for marriage and my family 4.76923 High social status Job-major match 4.76924 Good prospect of promotion Stable and secure job 4.75025 Job matches interests and personality Job with high social approval 4.70626 Taking care of parents Job manifests my values 4.64327 High social status Stable and secure job 4.62528 High social status Taking care of parents 4.61529 Good prospect of promotion Work independently and autonomously 4.61130 Stable and secure job Job matches interests and personality 4.563

References

Bailey, J.J., & Kinerson, C. (2005). Regret avoidance and risk tolerance. Financial Counseling and Planning, 16(1), 23–28.Blustein, D.L., Devenis, L.E., & Kidney, B.A. (1989). Relationship between the identity formation process and career development. Journal of Counseling Psychology, 36(2),

196–202.Chen, P. -H. (2009). A counseling model for self-relation coordination for Chinese clients with interpersonal conflicts. The Counseling Psychologist, 37(7), 987–1009.

http://dx.doi.org/10.1177/0011000009339975.Chung, T., & Mallery, P. (1999). Social comparison, individualism–collectivism, and self-esteem in china and the United States. Current Psychology, 18(4), 340–352.

http://dx.doi.org/10.1007/s12144-999-1008-0.

19X. Li et al. / Journal of Vocational Behavior 86 (2015) 10–19

Drugowitsch, J., Moreno-Bote, R., Churchland, A.K., Shadlen, M.N., & Pouget, A. (2012). The cost of accumulating evidence in perceptual decision making. The Journal ofNeuroscience, 32(11), 3612–3628.

Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117–140.Fretz, B.R., & Leong, F.T. (1982). Career development status as a predictor of career intervention outcomes. Journal of Counseling Psychology, 29, 388–393.Gati, I. (2013). Advances in career decision making. In B. Walsh, M.L. Savickas, & P. Hartung (Eds.), Handbook of vocational psychology: Theory, research, and practice

(pp. 183–215) (4th ed.). New York, NY: Routledge.Gottfredson, L. (1996). Gottfredson's theory of circumscription and compromise. In D. Brown, & L. Brooks & Associates (Eds.), Career choice and development

(pp. 179–232). San Francisco, CA: Jossey-Bass Publishers.Hayes, A.F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408–420.He, G., & Bai, F. (1997). The effect of reference point on decision making under risk. Acta Psychologica Sinica, 29(2), 178–186.Heine, S.J., & Hamamura, T. (2007). In search of East Asian self-enhancement. Personality and Social Psychology Review, 11(4), 4–27.Hoelzl, E., & Loewenstein, G. (2005). Wearing out your shoes to prevent someone else from stepping into them: anticipated regret and social takeover in sequential

decisions. Organizational Behavior and Human Decision Processes, 98, 15–27.Holland, J.L. (1997). Making vocational choices: a theory of vocational personalities and work environment (3rd ed.). Odessa FL: Psychological Assessment Resources.Holland, J.L., Gottfredson, D.C., & Power, P.G. (1980). Some diagnostic scales for research in decision making and personality: Identity, information, and barriers. Journal

of Personality and Social Psychology, 39(6), 1191–1200.Holland, J.L., Johnston, J.A., & Asama, N.F. (1993). The vocational identity scale: A diagnostic and treatment tool. Journal of Career Assessment, 1(1), 1–12.Hou, Z. -J. (2002). The influence of parental career expectations of children and parent–child relationship on the career aspiration and decision-making of middle school stu-

dents in Beijing. PhD. the Chinese University of Hongkong.Hou, Z. -J., Chen, S. -F., Zhou, S. -L., & Li, X. (2012). Development of parental career expectations scale for college students. Chinese Journal of Clinical Psychology, 20,

593–596.Jin, S. -H., & Li, X. (2005). The structure of college students's work values: intentions and instruments. Acta Psychological Sinica, 37, 650–657.Kahneman, D. (2003). A perspective on judgment and choice: Mapping bounded rationality. American Psychologist, 58, 697–720.Kidd, J.M. (1998). Emotion: An absent presence in career theory. Journal of Vocational Behavior, 52(3), 275–288. http://dx.doi.org/10.1006/jvbe.1997.1629.Kim, D., Pan, Y., & Park, H.S. (1998). High-versus low-context culture: A comparison of Chinese, Korean, and American cultures. Psychology and Marketing, 15(6),

507–521.Krieshok, T.S., Black, M.D., & McKay, R.A. (2009). Career decision making: The limits of rationality and the abundance of non-conscious processes. Journal of Vocational

Behavior, 75(3), 275–290. http://dx.doi.org/10.1016/j.jvb.2009.04.006.Kuhl, J. (1992). A theory of self-regulation: Action versus state orientation, self-discrimination, and some applications. Applied Psychology, 41(2), 97–129.Kumar, P. (2004). The effects of social comparison on inaction inertia. Organizational Behavior and Human Decision Processes, 95, 175–185.Li, X., Hou, Z. -J., & Feng, M. (2013). Relationships between parental career expectation, proactive personality, career adaptability, and career decision-making difficulty

among Chinese College Students. Chinese Journal of Clinical Psychology, 21(2), 263–267.Linde, J., & Sonnemans, J. (2012). Social comparison and risky choices. Journal of Risk and Uncertainty, 44(1), 45–72. http://dx.doi.org/10.1007/s11166-011-9135-z.Marcia, J.E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3(5), 551–558. http://dx.doi.org/10.1037/h0023281.Markus, H.R., & Kitayama, S. (1991). Culture and the self: implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224–253.McLean, E.R., Bryan, N.B., Tanner, J.R., & Smits, S.J. (1993). The structure of job attitudes among entry-level I/S professionals: A path-analytic analysis. Paper presented at

the conference on computer personnel research NY, USA.Osipow, S.H. (1983). Theories of career development (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall.Osipow, S.H., Carney, C.G., Winer, J.L., Yanico, B., & Koschier, M. (1976). The career decision scale. Columbus, OH: Marathon Consulting and Press.Parsons, E. (1909). Choosing vocation. Boston: Houghton-Mifflin.Peters, E., Vastfjall, D., Garling, T., & Slovic, P. (2006). Affect and decision making: A “hot” topic. Journal of Behavioral Decision Making, 19, 79–85.Porfeli, E.J., Lee, B., Vondracek, F.W., &Weigold, I.K. (2011). Amulti-dimensional measure of vocational identity status. Journal of Adolescence, 34(5), 853–871. http://dx.

doi.org/10.1016/j.adolescence.2011.02.001.Robinson, D.A., & Cooper, S.E. (1998). The relationship of Kohut's self-psychology to career choice certainty and satisfaction. Journal of College Student Development,

29(3), 228–232.Saka, N., & Gati, I. (2007). Emotional and personality-related aspects of persistent career decision-making difficulties. Journal of Vocational Behavior, 71(3), 340–358.Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D.R. (2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of

Personality and Social Psychology, 83(5), 1178–1197.Schwarz, N., & Clore, G.L. (1988). How do I feel about it? The informative function of affective states. In K. Fiedler, & I. Forgas (Eds.), Affect, cognition, and social behavior

(pp. 44–62). Göttingen: Hogrefe.Super, D.E. (1980). A life-span, life-space approach to career development. Journal of Vocational Behavior, 16(3), 282–298. http://dx.doi.org/10.1016/0001-8791(80)

90056-1.Tracey, T.J.G., & Darcy, M. (2002). An idiothetic examination of vocational interests and their relation to career decidedness. Journal of Counseling Psychology, 49(4),

420–427.Whiston, S.C., & Keller, B.K. (2004). The influences of the family of origin on career development: A review and analysis. The Counseling Psychologist, 32, 493–568.

http://dx.doi.org/10.1177/0011000004265660.White, J.B., Langer, E.J., Yariv, L., & Welch, J. C., IV (2006). Frequent social comparisons and destructive emotions and behaviors: The dark side of social comparisons.

Journal of Adult Development, 13(1), 36–44.White, K., & Lehman, D.R. (2005). Culture and social comparison seeking: The role of self-motives. Personality and Social Psychology Bulletin, 31(2), 232–242. http://dx.

doi.org/10.1177/0146167204271326.Zeelenberg, M. (1999). Anticipated regret, expected feedback and behavioral decision making. Journal of Behavioral Decision Making, 12(2), 93–106.Zhou, A.B., & Ma, X.F. (2007). The compromise strategy of university students' career decision-making. Psychological Science, 30(5), 1247–1249.